A control system of a shaping machine for construction waste recycled aggregate

Abstract

The application relates to the technical field of device control, and discloses a control system of a shaping machine for construction waste recycled aggregate, which comprises: a collection subsystem for collecting a target load current of a main driving motor in the shaping machine and identifying a material component type according to the target load current; a calculation subsystem for selecting an optimization coefficient corresponding to the material component type and calculating a target gap based on the optimization coefficient; a control subsystem for generating adjustment error data based on the target gap and a current gap, controlling a hydraulic cylinder to adjust the position of a back plate, and generating a first gap adjustment instruction; and a correction subsystem for detecting motor temperature rise data, predicting a hydraulic response delay, correcting the first gap adjustment instruction based on the hydraulic response delay, and obtaining a second gap adjustment instruction. The application effectively solves the problems of overshoot and oscillation in traditional control, realizes early warning and active prevention of the risk of material blockage, and improves the continuous operation capacity and reliability of the shaping machine.

Description

Technical Field

[0001] This invention relates to the field of equipment control technology, and in particular to a control system for a shaping machine for recycled aggregates from construction waste. Background Technology

[0002] Construction waste aggregate shaping machines are key equipment for the resource utilization of construction waste. Traditional shaping machines use fixed gaps and uniform control parameters, which cannot be adjusted according to the characteristics of the materials, resulting in unstable shaping effects, excessive energy consumption, and accelerated equipment wear. Existing control systems lack the ability to identify the composition of construction waste in real time, cannot distinguish the impact differences of materials with different hardness on the equipment, and are difficult to effectively separate the interference effects of lightweight materials. In particular, when the material hardness changes abruptly, it is prone to control oscillations, affecting the shaping quality and equipment stability. Summary of the Invention

[0003] This invention provides a control system for a shaping machine for recycled aggregates from construction waste. This invention effectively solves the problems of overshoot and oscillation in traditional control, realizes early warning and proactive prevention of material blockage risk, and improves the continuous operation capability and reliability of the shaping machine.

[0004] In a first aspect, the present invention provides a control system for a shaping machine for recycled aggregates from construction waste, the control system comprising: The acquisition subsystem is used to acquire the target load current of the main drive motor in the shaping machine and identify the material composition type based on the target load current; A calculation subsystem is used to select the optimization coefficient corresponding to the material composition type and calculate the target gap based on the optimization coefficient; The control subsystem is used to generate adjustment error data based on the target gap and the current gap, and to control the hydraulic cylinder to adjust the position of the counterattack plate according to the adjustment error data, thereby generating a first gap adjustment command. The correction subsystem is used to detect motor temperature rise data and predict hydraulic response delay, and correct the first gap adjustment command based on the hydraulic response delay to obtain the second gap adjustment command.

[0005] In conjunction with the first aspect, in a first implementation of the first aspect of the present invention, the acquisition subsystem further includes: The acquisition module is used to acquire the instantaneous load current of the main drive motor in the shaping machine; The identification module is used to measure the duration of the peak signal and the current drop slope in the instantaneous load current, identify and eliminate interference signals from lightweight materials, and obtain the target load current. The analysis module is used to divide the target load current by the motor rated current to obtain the load current ratio. The identification module is used to identify the material composition type as concrete block when the load current ratio is greater than a first target value; to identify the material composition type as waste brick when the load current ratio is between a second target value and a first target value; and to identify the material composition type as stone when the load current ratio is less than a second target value, wherein the first target value is greater than the second target value.

[0006] In conjunction with the first aspect, in a second implementation of the first aspect of the present invention, the identification module is specifically used for: Identify the peak segments in the instantaneous load current that exceed the average load current, and measure the time span from the start of the rise to the end of the fall for each peak segment to obtain the duration; At the end of each peak period, calculate the rate of change of the load current from the peak to the average value to obtain the current drop slope. The duration is compared with the time discrimination threshold, and the current drop slope is compared with the slope discrimination threshold. When the duration is shorter than the time discrimination threshold and the current drop slope exceeds the slope discrimination threshold, it is marked as a lightweight material interference signal. The current data marked as interference signals of lightweight materials are removed from the instantaneous load current, and the effective load characteristics corresponding to heavy construction waste are retained to obtain the target load current.

[0007] In conjunction with the first aspect, in a third implementation of the first aspect of the present invention, the computing subsystem is specifically used for: According to the material composition type, the corresponding optimization coefficient is retrieved from the preset coefficient library, where concrete blocks correspond to the first set of coefficients, waste bricks correspond to the second set of coefficients, and stone corresponds to the third set of coefficients. The current load current of the shaping machine and the residence time of the material in the cavity are collected synchronously to obtain the current operating status parameters; The optimization coefficients are used to construct a relationship function between the counter-attack plate gap and the shaping efficiency and energy consumption. The current operating state parameters are substituted into the relationship function to calculate the response value. The extreme points are determined based on the response values, and the target gap is calculated based on the extreme points to maximize the shaping efficiency while minimizing energy consumption.

[0008] In conjunction with the first aspect, in a fourth implementation of the first aspect of the present invention, the control subsystem further includes: The judgment module is used to subtract the current gap of the counterattack plate from the target gap to obtain the gap deviation, and at the same time determine the absolute value range of the gap deviation to obtain adjustment error data; The selection module is used to select the corresponding current PID parameters based on the adjustment error data. The generation module is used to generate hydraulic adjustment commands based on the adjustment error data and the current PID parameters; The verification module is used to verify whether the hydraulic adjustment command satisfies the stability condition that the product of the gap change gradient and the load change gradient is less than zero. When the stability condition is met, the hydraulic adjustment command is used as the first gap adjustment command and output.

[0009] In conjunction with the first aspect, in a fifth implementation of the first aspect of the present invention, the generation module is specifically used for: The adjustment error data is multiplied by the proportional coefficient in the current PID parameters to obtain the proportional control component. The adjustment error data is integrated over time and then multiplied by the integral coefficient to obtain the integral control component. The adjustment error data is differentiated over time and then multiplied by the derivative coefficient to obtain the derivative control component. The proportional control component, the integral control component, and the derivative control component are proportionally converted to obtain the target hydraulic speed. When the target hydraulic speed exceeds a preset speed threshold, the hydraulic cylinder's operating speed is limited to the preset speed threshold; when the target hydraulic speed does not exceed the preset speed threshold, the target hydraulic speed remains unchanged, thus obtaining the limited hydraulic speed. Hydraulic adjustment commands are generated based on the restricted hydraulic speed.

[0010] In conjunction with the first aspect, in a sixth implementation of the first aspect of the present invention, the correction subsystem further includes: The acquisition module is used to measure the difference between the current temperature of the motor and the initial temperature at startup to obtain the motor temperature rise data, and at the same time, to determine the current wear status of the impact plate based on the motor temperature rise data; The calculation module is used to calculate the wear gap deviation of the impact plate based on the current wear state; The prediction module is used to analyze the real-time change rate of the target load current, multiply the real-time change rate by a preset prediction coefficient and a delay time constant to predict the time difference between the hydraulic cylinder receiving the first gap adjustment command and completing the action, and obtain the hydraulic response delay. The compensation module is used to add the wear gap deviation to the first gap adjustment command to compensate for the wear effect, and then adjust the command timing in advance according to the hydraulic response delay to obtain the second gap adjustment command.

[0011] In conjunction with the first aspect, in the seventh implementation of the first aspect of the present invention, the calculation module is specifically used for: Calculate the percentage of wear exceeding the standard wear threshold compared to the current wear state; The first wear compensation coefficient is adjusted based on the wear exceeding the standard ratio to obtain the second wear compensation coefficient; Accumulate the load current values ​​of the shaping machine from its first start to the current moment for all operating periods, and continuously integrate the load current values ​​to record the total impact load borne by the shaping machine, thereby obtaining the historical integrated value of the load current. The wear gap deviation caused by cumulative impact wear of the impact plate is quantified by multiplying the second wear compensation coefficient by the historical integral value of the load current.

[0012] In conjunction with the first aspect, in the eighth implementation of the first aspect of the present invention, the control system of the forming machine for recycled aggregate from construction waste further includes a blockage risk analysis subsystem, which includes: The differential operation module is used to perform two consecutive differential operations on the target load current to obtain the second derivative. The processing module is used to collect load current fluctuation data, calculate the standard deviation and root mean square deviation of the load current fluctuation data, and calculate the load current variance value based on the standard deviation and root mean square deviation. The synthesis module is used to synthesize the second derivative and the load current variance value into a blockage precursor identification parameter, and generate a blockage risk value based on the blockage precursor identification parameter; The risk control module is used to increase the gap value in the second gap adjustment command to a safe gap value when the material blockage risk value exceeds the risk threshold, and to keep the second gap adjustment command unchanged when the material blockage risk value does not exceed the risk threshold, and finally obtain the third gap adjustment command.

[0013] In conjunction with the first aspect, in the ninth implementation of the first aspect of the present invention, the synthesis module is specifically used for: The sensitivity weighting coefficient of the second derivative and the fluctuation weighting coefficient of the load current variance are determined based on historical material blockage events. Multiplying the second derivative by the sensitivity weighting coefficient yields the first blockage parameter, and multiplying the load current variance by the fluctuation weighting coefficient yields the second blockage parameter. Construct positive and negative sample pairs for the twin support vector machine, use the identification parameters during historical normal operation as negative samples, and use the identification parameters during historical material blockage precursors as positive samples, train the twin support vector machine to learn the feature boundaries between the material blockage state and the non-material blockage state, and obtain the material blockage identification classifier; The first and second blockage parameters are used as blockage precursor identification parameters and input into the blockage identification classifier. The similarity distance between the blockage precursor identification parameters and positive samples and the similarity distance with negative samples are calculated, and the blockage risk value is output.

[0014] The technical solution provided by this invention achieves real-time automatic identification of construction waste composition through motor load current characteristic analysis. It can accurately distinguish materials of different hardness, such as concrete blocks, waste bricks, and stone, and effectively eliminate interference from lightweight impurities such as insulation materials and plastics, solving the technical problem of traditional technologies being unable to identify material differences. Based on the dynamic selection of dedicated optimization coefficients according to material composition type, it achieves the optimal balance between shaping efficiency and energy consumption through a three-dimensional relationship function. It automatically adjusts the impact plate gap according to the characteristics of different construction wastes, significantly improving the stability and consistency of the shaping effect compared to traditional fixed-gap control methods. Employing a segmented PID control strategy, it automatically switches control parameters according to the gap error magnitude, and uses a dedicated stability criterion to prevent positive feedback oscillation between the gap and load, achieving a dynamic control mode with fast response, balanced adjustment, and fine control, effectively solving the overshoot and oscillation problems in traditional control. By using motor temperature rise data to evaluate the wear state of the impact plate in real time, it dynamically adjusts the wear compensation coefficient, and accurately calculates the gap deviation caused by wear based on the equipment's full life-cycle load history data, ensuring the stability of control accuracy during long-term operation and overcoming the technical deficiency of traditional technologies that ignore the impact of wear. Based on the prediction of hydraulic system response delay using the load current change rate, the command timing is adjusted in advance for compensation and correction, eliminating the impact of system lag on control accuracy and achieving more precise and real-time gap adjustment response. A blockage precursor identification system is constructed using the second derivative of the load current and variance analysis, enabling early warning and proactive prevention of blockage risks, thus improving the continuous operation capability and reliability of the forming machine. Attached Figure Description

[0015] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic diagram of the control system of the shaping machine for recycled aggregate from construction waste in an embodiment of the present invention. Detailed Implementation

[0017] This invention provides a control system for a shaping machine for recycled aggregate from construction waste. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, system, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, systems, products, or devices.

[0018] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 One embodiment of the control system for the shaping machine of recycled aggregate from construction waste in this invention includes: The data acquisition subsystem is used to acquire the target load current of the main drive motor in the shaping machine and identify the material composition type based on the target load current. Specifically, the acquisition subsystem uses an acquisition module to collect the instantaneous load current of the main drive motor of the shaping machine at high frequency, capturing subtle current changes during material impact and processing. The identification module analyzes the peak values ​​in the instantaneous load current signal, measures the duration of the peak signal, and calculates the current drop slope after the peak. By comparing the duration with a preset time threshold and combining the numerical characteristics of the drop slope, the interference characteristics of lightweight materials during the crushing process are identified, as lightweight materials generate peak signals with extremely short durations and large drop slopes when in contact with the impact plate. After detecting such interference, the identification module removes abnormal peaks to obtain the target load current. The analysis module calculates the ratio of the target load current to the rated current of the motor, obtaining the load current ratio. The load current ratio is input into the second identification module, which determines the material composition type based on preset first and second target values. When the load current ratio is greater than the first target value, the material composition type is determined to be concrete block; when the ratio is between the second and first target values, it is determined to be waste brick; when the ratio is less than the second target value, it is determined to be stone. The first target value is greater than the second target value.

[0019] The calculation subsystem is used to select the optimization coefficient corresponding to the material composition type and calculate the target gap based on the optimization coefficient; Specifically, based on the material composition, corresponding optimized coefficients are retrieved from a pre-set coefficient library. Concrete blocks are associated with the first set of coefficients, waste bricks with the second set, and stone with the third set. Each set of coefficients is obtained through experimental calibration and fitting with historical operating data, reflecting the energy consumption sensitivity and gap change response characteristics of materials with different hardness during the shaping process. The current load current of the shaping machine and the actual residence time of the material in the cavity are simultaneously collected. These two parameters constitute the key inputs to the current operating status, reflecting the force state of the impact plate and the degree to which the material is subjected to force in the cavity. The calculation subsystem combines the optimized coefficients with the current operating status parameters to construct a multivariate relationship function between the impact plate gap, shaping efficiency, and energy consumption. This multivariate relationship function can comprehensively consider the impact of gap changes on crushing and shaping effects, processing efficiency, and energy consumption within the same model. Substituting the current load current and material residence time into the multivariate relationship function yields a response value that simultaneously reflects the comprehensive performance of efficiency and energy consumption. The trend of the response value is analyzed using mathematical solution methods to find the corresponding extreme points. At the extreme point, the shaping efficiency is at its peak, while energy consumption is reduced to its minimum. The calculation subsystem uses this extreme point as a basis to determine the target gap.

[0020] The control subsystem is used to generate adjustment error data based on the target gap and the current gap, and to control the hydraulic cylinder to adjust the position of the counterattack plate according to the adjustment error data, thereby generating the first gap adjustment command. Specifically, the judgment module calculates the difference between the target gap value and the current gap of the counter-attack plate to obtain the gap deviation, representing the difference between the current gap and the ideal setting. The absolute value of the gap deviation is taken, and its numerical range is determined to generate adjustment error data, used to distinguish between working states requiring rapid response, large adjustments, or only fine-tuning. Based on the adjustment error data, the selection module chooses the most suitable set of parameters from a set of preset PID control parameters. For example, a PID parameter set with high response speed is selected when the deviation is large; a balanced parameter set balancing response and stability is selected when the deviation is medium; and a low-gain parameter set for fine adjustment is selected when the deviation is small. The generation module inputs the adjustment error data and the currently selected PID parameters into the PID calculation model to generate the corresponding hydraulic adjustment command. The hydraulic adjustment command directly controls the action of the hydraulic cylinder to drive the counter-attack plate to adjust towards the target gap. To ensure the stability of the adjustment process, the verification module verifies the hydraulic adjustment command before outputting it, calculating the product of the gap change gradient and the load change gradient, and determining whether the product is less than zero. When the product is negative, it indicates that the gap change trend is opposite to the load change trend, that is, the system is in a stable adjustment state. At this time, the test module allows the hydraulic adjustment command to be used as the first gap adjustment command and output to the actuator.

[0021] The correction subsystem is used to detect motor temperature rise data and predict hydraulic response delay. Based on the hydraulic response delay, the first gap adjustment command is corrected to obtain the second gap adjustment command.

[0022] Specifically, the acquisition module performs real-time temperature monitoring of the motor. By measuring the difference between the current motor temperature and the initial temperature at startup, it calculates the motor's temperature rise data, reflecting the motor's operating status under current load conditions. This data also serves as an indirect indicator of the wear degree of the impact plate, as the wear of the impact plate is accompanied by increased motor energy consumption and accelerated temperature rise during long-term high-load operation of the forming machine. The current wear state of the impact plate is determined based on the correlation between the temperature rise and historical calibration data, and the result is transmitted to the calculation module. The calculation module calculates the gap deviation value caused by the wear of the impact plate, reflecting the difference between the current actual gap and the ideal set value due to wear. The prediction module performs real-time rate-of-change analysis on the target load current. By calculating the speed and direction of change, combined with preset prediction coefficients and delay time constants, it calculates the time difference between the hydraulic cylinder receiving the first gap adjustment command and actually completing the action, thus obtaining the hydraulic response delay. The compensation module adds the wear clearance deviation to the first clearance adjustment command to compensate for the problem of insufficient or excessive clearance caused by the wear of the counterattack plate. At the same time, it adjusts the execution sequence of the command in advance according to the hydraulic response delay, so that the actuator can complete the position change at the correct time in actual operation, and generates and outputs the second clearance adjustment command after double correction.

[0023] In one specific embodiment, the acquisition subsystem further includes: The acquisition module is used to acquire the instantaneous load current of the main drive motor in the shaping machine; The identification module is used to measure the duration of the peak signal and the current drop slope in the instantaneous load current, identify and eliminate interference signals from lightweight materials, and obtain the target load current. The analysis module is used to divide the target load current by the motor rated current to obtain the load current ratio. The identification module is used to identify the material composition type as concrete block when the load current ratio is greater than the first target value; to identify the material composition type as waste brick when the load current ratio is between the second target value and the first target value; and to identify the material composition type as stone when the load current ratio is less than the second target value, wherein the first target value is greater than the second target value.

[0024] Specifically, the acquisition module installs a high-sensitivity current sensor in the input circuit of the main drive motor of the shaping machine, with a sampling frequency reaching millisecond levels to ensure complete capture of the current fluctuation characteristics generated when materials enter the crushing chamber and collide with the impact plate. Because construction waste contains components of varying hardness and density, such as concrete blocks, waste bricks, and stone, and is mixed with lightweight materials like wood, plastic sheets, and foam blocks, these lightweight materials generate instantaneous high-amplitude but extremely short-duration current peak signals during impact, easily leading to misjudgment of material type. The identification module performs peak analysis on the instantaneous load current signal, detecting each peak segment significantly higher than the background load, and measuring the duration of each peak segment—the time from the peak rise point to the baseline or stable segment. Simultaneously, it calculates the slope of the peak decline segment, reflecting the speed at which the peak disappears. Since contact with lightweight materials ends in a very short time, causing the current to drop rapidly, the decline slope of lightweight materials is significantly greater than that of high-density, hard materials. The identification module compares the duration and descent slope with a preset threshold for lightweight material interference. When the duration of a peak is less than the set threshold and the descent slope is greater than the specified threshold, the peak signal is determined to be an interference signal caused by lightweight materials, and the interference signal is removed from the dataset. The remaining signal segment is the target load current that accurately reflects the material impact characteristics and crushing load characteristics. The analysis module standardizes the target load current by dividing the target load current value by the rated current of the motor to obtain a dimensionless load current ratio. The load current ratio is input into a second identification module, which has built-in material type determination logic and sets a first target value and a second target value, with the first target value being greater than the second target value. When the load current ratio is greater than the first target value, the material composition is determined to be concrete block. This is because concrete blocks have high hardness and density, requiring a large impact force to crush, resulting in a significantly higher load current for the main drive motor when processing such materials. When the load current ratio is between the second and first target values, the material composition is determined to be waste brick. Waste brick is lower in hardness and density than concrete blocks, but still has a certain strength, so its corresponding motor load will be higher than that of stone but lower than that of concrete blocks. When the load current ratio is less than the second target value, the material composition is determined to be stone. This is mainly because stone materials, with their regular shape and low crushing resistance, have a limited impact on the additional load of the motor during recycled aggregate processing, resulting in a lower current ratio.

[0025] In one specific embodiment, the identification module is specifically used for: Identify the peak segments in the instantaneous load current that exceed the average load current, and measure the time span from the start of the rise to the end of the fall of each peak segment to obtain the duration; At the end of each peak period, calculate the rate of change of the load current from the peak to the average value to obtain the current drop slope. The relationship between the duration and the time discrimination threshold is compared, and the relationship between the current drop slope and the slope discrimination threshold is also compared. When the duration is shorter than the time discrimination threshold and the current drop slope exceeds the slope discrimination threshold, it is marked as a lightweight material interference signal. The current data marked as interference signals of lightweight materials are removed from the instantaneous load current, and the effective load characteristics corresponding to heavy construction waste are retained to obtain the target load current.

[0026] Specifically, peak segment detection marks the starting point of the rise when the signal crosses above the mean, and marks the ending point of the fall when the signal falls back and crosses below the mean, thus determining the complete peak segment. To prevent multiple false entries and exits caused by high-frequency glitches, peak detection introduces up and down travel hysteresis bands. The rise judgment uses "mean plus a small margin," and the fall judgment uses "mean minus a small margin," and a minimum inter-peak interval is set to ensure that each collision forms only one valid peak segment. The duration is directly measured by the time span from the rising point to the falling point. At the end of each peak segment, the processing link uses the difference between the peak amplitude and the mean amplitude as the fall amplitude, and uses the fall time from the peak to the mean as the time scale to calculate the rate of change of the current from the peak to the mean, thus obtaining the current fall slope. Lightweight debris such as wood, foam, or thin plastic sheets have low inertia, short impact time, and rapid rebound during impact, forming a current pattern of "extremely short duration and extremely steep fall." Heavy construction waste such as concrete blocks, waste bricks, and stone have more thorough contact and greater crushing resistance during impact, resulting in a longer peak duration and a gentler fall. The threshold joint discrimination process adopts a strategy of comparing time thresholds and slope thresholds simultaneously to reduce the risk of false alarms caused by a single threshold. The time discrimination threshold is set to 0.1 seconds based on the calibration results of the working conditions to distinguish between extremely short contact events and effective crushing events. The slope discrimination threshold is calibrated experimentally based on the equipment power level and hydraulic stiffness, and is adaptively fine-tuned by combining the current variance within a 5-second window. When the volatility increases significantly, the threshold is raised accordingly to suppress non-contact false peaks caused by overall working condition disturbances. The algorithm traverses each peak segment one by one. If the duration is shorter than the time discrimination threshold and the current drop slope exceeds the slope discrimination threshold, the peak segment is marked as a lightweight material interference signal. The marked interference segments are reconstructed using smooth interpolation or neighborhood steady-state substitution strategies. Linear or spline interpolation is performed using the mean neighborhood samples at both ends of the segment to ensure the reconstructed waveform remains continuous in both time and amplitude; alternatively, the moving average of the steady-state conditions within the neighborhood is used to backfill the samples within the segment, maintaining consistency in power estimation. An amplitude suppression coefficient (engineering parameter 0.6) is introduced during the reconstruction stage to further suppress residual short spikes, reducing their impact on subsequent ratio determination. After reconstruction, the moving average and standard deviation of the entire time series are recalculated to correct for slight baseline shifts caused by removal. Considering complex situations such as peak overlap and rapid bursts, the peak tracker maintains a unique identifier for each segment during detection and performs merging when the distance between the start and end points of adjacent segments is less than the minimum inter-peak interval, avoiding the misclassification of a single heavy impact as multiple short events. During sudden changes in operating conditions or the early stages of material blockage, the window variance will increase significantly. Variance information is used as a bypass signal for joint discrimination. If necessary, the slope threshold is temporarily increased to prevent the overall load step change from being regarded as a minor disturbance.For high-frequency spikes caused by extreme noise or sensor jitter, median filtering and minimum duration pre-screening are superimposed at the front end to weaken non-physical spikes at the signal source. The retained waveform represents the stress characteristics of heavy construction waste, thereby forming the target load current.

[0027] The identification of material composition types as concrete blocks, waste bricks, or stone includes: establishing a theoretical mathematical model for identifying the load characteristics of construction waste; constructing a piecewise continuous identification function using the Heaviside step function, where concrete blocks, waste bricks, and stone correspond to different step trigger conditions; achieving clear classification of components through the jumping characteristics of the step function, thus obtaining the step function identification model; calculating the stress response thresholds of different construction waste materials; determining the stress thresholds for concrete blocks as 45 MPa, waste bricks as 25 MPa, and stone as 80 MPa through material mechanics analysis; converting the stress thresholds into corresponding motor load current thresholds to obtain the step trigger conditions; and designing a smoothing scheme for the Heaviside step function. The processing algorithm uses the Sigmoid function to approximate the step function and eliminate discontinuities. By adjusting the smoothing parameter, the transition characteristics of the function near the threshold are controlled to avoid recognition jitter caused by hard switching, resulting in a smooth step recognition function. A weighted superposition algorithm for multiple step functions is constructed. When the load current triggers multiple step conditions simultaneously, the dominant material component is determined through a weight allocation mechanism. A priority sequence of step functions and a conflict resolution strategy are established to obtain the step function decision matrix. An adaptive adjustment mechanism for the step function parameters is established. The stress threshold and smoothing parameter are dynamically optimized according to the actual recognition effect. The recognition error is minimized through the gradient descent algorithm, making the step function model adaptable to the differences in material properties of construction waste in different regions, resulting in an adaptive step recognition system.

[0028] In one specific embodiment, the computing subsystem is specifically used for: The corresponding optimization coefficients are retrieved from the preset coefficient library according to the material composition type. Concrete blocks correspond to the first set of coefficients, waste bricks correspond to the second set of coefficients, and stone corresponds to the third set of coefficients. The current load current of the shaping machine and the residence time of the material in the cavity are collected synchronously to obtain the current operating status parameters; The relationship function between the counter-attack plate gap and the shaping efficiency and energy consumption is constructed using optimization coefficients. The current operating state parameters are substituted into the relationship function to calculate the response value. The extreme points are determined based on the response values, and the target gap is calculated based on these extreme points to maximize the shaping efficiency while minimizing energy consumption.

[0029] Specifically, optimized coefficient sets are extracted from a pre-set coefficient library based on the material composition type, ensuring that the coefficients match the material hardness. The coefficient library is calibrated according to three typical types of construction waste: concrete blocks use the first set of coefficients (k1=2.5, k2=1.8, k3=0.6); waste bricks use the second set (k1=2.0, k2=1.5, k3=0.8); and stone uses the third set (k1=1.5, k2=1.2, k3=1.0). The current load current of the shaping machine and the material residence time within the crushing chamber are simultaneously collected to form a set of current operating status parameters. The load current is sampled at 1000Hz and used with a 5-second sliding window for mean and variance to suppress noise and enhance robustness, while retaining instantaneous changes to reflect impact intensity. Residence time is estimated jointly by feed rate measurement, rotor speed, and a chamber flow model, or obtained from upstream metering and image timing. These two parameters, along with the impact plate gap, characterize the processing intensity and energy consumption level, and are key independent variables in the relational function. When constructing the relationship function between the impact plate gap, forming efficiency, and energy consumption, the calculation subsystem merges the three types of influences into a single response value for extreme value solving. The relationship function uses the effect of gap G on forming efficiency, the indicator effect of load current on energy consumption, and the decay effect of residence time on forming sufficiency as three main lines, and weights these three main lines using coefficients from the first, second, or third group. k1 reflects the diminishing marginal benefit of gap change on forming quality, k2 reflects the logarithmic energy consumption sensitivity of load current relative to rated current, and k3 reflects the exponential decay effect of residence time with time constant τ. The sum of these three forms a single-peak comprehensive response value. To enhance engineering applicability, operating condition constraints are added to the outer layer of the function, such as allowable upper and lower limits for gap, hydraulic mechanism stroke limits, and penalty terms related to material blockage risk. When the standard deviation of motor load fluctuation or the absolute value of the motor load change rate increases abnormally, a penalty is imposed on excessively small gaps, thereby avoiding sacrificing stability for nominal efficiency under extreme loads. Substituting the current load current and residence time into the relational function yields the response curve for the current operating condition. Since the response value exhibits a unimodal structure with respect to the gap, two complementary methods are employed for extreme value determination: a fast analytical solution is used to locate the peak point when the analytical expression is clear and the coefficients are stable; a numerical method, such as interval ternary search or a one-dimensional Newton search with boundaries, is used when the operating condition disturbance is large or after adding penalty and constraint terms to find the maximum response value within the set feasible region of the gap. During the solution process, the gradient sign change and convergence step size are monitored simultaneously to avoid out-of-bounds errors caused by numerical fluctuations. To adapt to the actual achievable accuracy of the hydraulic actuator, a 1–2 mm fine-tuning search window is introduced near the theoretical peak value. Fine-grained scanning confirms the center position of the response platform, thereby offsetting actuator quantization and hysteresis errors. The target gap is calculated based on the extreme point. The operating condition corresponding to the target gap maximizes the shaping efficiency while minimizing energy consumption.The calculation subsystem performs time filtering or exponential smoothing on the target gap and the optimal solution of the previous sampling period to avoid set value jitter due to instantaneous disturbances; at the same time, it applies a small protection band to the target gap by combining the sliding window variance, which restricts overly aggressive gap reduction actions during the rising phase of the window variance, and finally obtains the target gap.

[0030] The process involves constructing a three-dimensional relationship function between the impact plate gap, shaping efficiency, and energy consumption using optimization coefficients. This includes: establishing a three-dimensional composite function of impact plate gap, shaping efficiency, and energy consumption, containing an inverse square root term for the gap reflecting diminishing returns, a logarithmic term for the load ratio reflecting the logarithmic characteristics of the load response, and an exponential term for the residence time ratio reflecting the exponential decay of residence time, resulting in a three-term composite optimization function; designing a physical model for the inverse square root term in the composite function, deriving the inverse square root relationship between the gap and shaping efficiency based on particle collision dynamics theory, and verifying through finite element simulation that the shaping effect is optimal when the exponential parameter of the inverse square root term is -0.5, thus obtaining the inverse square root term parameter; and constructing a load response characteristic model for the logarithmic term, based on the motor load... The nonlinear response relationship between load and material hardness was investigated. A logarithmic response model for the load ratio was determined using the Weber-Fechner law, and the influence of the logarithmic base selection on optimization accuracy was analyzed to obtain the logarithmic term parameters. A time decay model for the exponential term was established. Based on the theory of material residence time distribution in the crushing chamber, the time constant of the exponential term was determined through a combination of Poisson and exponential distributions. The influence of different material compositions on the time constant was analyzed to obtain the exponential term parameters. A multi-objective optimization algorithm for a three-dimensional composite function was designed. A constrained optimization model was established using the Lagrange multiplier method, and the extreme points of the function were solved using a particle swarm optimization algorithm. This simultaneously satisfied the dual objectives of maximizing shaping efficiency and minimizing energy consumption, yielding the optimal gap analytical solution.

[0031] In one specific embodiment, the control subsystem further includes: The judgment module is used to subtract the current gap of the counterattack plate from the target gap to obtain the gap deviation, and at the same time determine the absolute value range of the gap deviation to obtain the adjustment error data. The selection module is used to select the corresponding current PID parameters based on the adjustment error data. The generation module is used to generate hydraulic adjustment commands based on the adjustment error data and the current PID parameters; The verification module is used to verify whether the hydraulic adjustment command meets the stability condition that the product of the gap change gradient and the load change gradient is less than zero. When the stability condition is met, the hydraulic adjustment command is used as the first gap adjustment command and output.

[0032] Specifically, the judgment module receives the target gap and compares it with the current gap of the counter-attack plate measured by the sensor link. The gap deviation is obtained by subtracting the two, and the absolute value of the gap deviation is used to classify the intervals, forming adjustment error data. The adjustment error data is categorized into three levels: when the absolute deviation is greater than 5mm, it is classified as fast response mode; when the absolute deviation is between 3mm and 5mm, it is classified as balanced response mode; and when the absolute deviation does not exceed 3mm, it is classified as fine response mode. Based on the adjustment error data, the selection module calls the corresponding parameter set from the preset PID parameter library. The fast response mode uses Kp=3.5, Ki=0.8, Kd=1.2; the balanced response mode uses Kp=2.2, Ki=0.5, Kd=0.9; and the fine response mode uses Kp=1.5, Ki=0.3, Kd=0.6. Kp, Ki, and Kd are three core parameters in the PID controller, representing the proportional coefficient, integral coefficient, and derivative coefficient, respectively. In each control cycle, the generation module inputs the adjustment error data and the current PID parameters into the controller kernel to calculate the target displacement and target speed of the hydraulic cylinder. At the same time, it superimposes a rate limiter and an amplitude limiter to protect the structural integrity of the hydraulic actuator and the counterattack plate. The maximum adjustment speed is limited to 2 mm / s. The displacement command introduces differential lead-ahead and anti-saturation integral clamping to avoid integral accumulation caused by large deviations over a long period of time and secondary overshoot when approaching the target. For scenarios where the error direction frequently reverses, the generation module activates a micro-zone dead zone and small-amplitude softening strategy to reduce the proportional gain proportionally within 1 mm, thereby reducing the probability of high-frequency back-and-forth adjustment. After the hydraulic adjustment command is generated, the verification module performs a double check based on stability constraints: It calculates the product of the gap change gradient and the load change gradient. Only when the product is less than zero is the gap change trend considered to be opposite to the motor load change trend, indicating the system is in the negative feedback zone and meets the stability condition. Furthermore, it imposes an amplitude constraint on the relationship between the instantaneous gap change and the moving average change of the motor load. The instantaneous gap change must be less than the product of the stability coefficient and the square root of the moving average change of the load. The stability coefficient is set to 0.15 to suppress over-adjustment during severe load fluctuations. If the product is negative and the amplitude constraint is met, the verification module confirms the hydraulic adjustment command as the first gap adjustment command and outputs it. If either condition is not met, anti-oscillation protection is triggered, immediately freezing the gap adjustment for 0.5 seconds and maintaining the current hydraulic holding pressure. Simultaneously, an instability event is recorded for adaptive tuning. After the command takes effect, the inspection module monitors the signs of the motor load change rate and the clearance change rate. When a trend of the same sign appears, it will quickly reduce the proportional gain and extend the integral time constant. If necessary, it will revert to the previous stable set point. When the material blockage risk assessment gives a medium to high risk, the inspection module will temporarily relax the output in the clearance increase direction and tighten the output in the clearance decrease direction according to the strategy to ensure that the load drops back to a safe level in a short time.

[0033] In one specific embodiment, the generation module is specifically used for: The proportional control component is obtained by multiplying the adjustment error data by the proportional coefficient in the current PID parameters. The integral control component is obtained by integrating the adjustment error data over time and multiplying it by the integral coefficient. The derivative control component is obtained by differentiating the adjustment error data over time and multiplying it by the derivative coefficient. The proportional control component, integral control component, and derivative control component are proportionally converted to obtain the target hydraulic speed. When the target hydraulic speed exceeds the preset speed threshold, the hydraulic cylinder's operating speed is limited to the preset speed threshold; when the target hydraulic speed does not exceed the preset speed threshold, the target hydraulic speed remains unchanged, thus obtaining the limited hydraulic speed. Hydraulic adjustment commands are generated based on the limited hydraulic speed.

[0034] Specifically, the proportional channel multiplies the adjustment error data by a proportional coefficient to obtain the proportional control component, converting instantaneous deviation into immediate correction force to quickly close the gap between the target gap and the current gap. The integral channel accumulates and sums the adjustment error data along the time axis, then multiplies it by the integral coefficient to obtain the integral control component. The integral channel eliminates long-term small steady-state errors, preventing the counter-attack plate from deviating from the target due to slight deviations. Simultaneously, the derivative channel calculates the rate of change of the adjustment error data over time and multiplies it by the derivative coefficient to obtain the derivative control component. The derivative component characterizes the error change trend and applies braking in advance when the error increases rapidly to suppress overshoot and oscillation. Considering the high-frequency noise of instantaneous load current, the derivative channel smooths the error signal with a small window or performs a first-order filter before calculating the rate of change, thus avoiding noise amplification. To suppress integral saturation, the integral channel sets upper and lower limits and pauses integral accumulation when the output speed reaches saturation to prevent secondary overshoot after the speed drops. The controller proportionally converts the proportional, integral, and derivative control components according to predetermined weights and maps them uniformly to the speed dimension on the hydraulic side to obtain the target hydraulic speed. The mapping coefficients are determined during the online calibration phase based on the effective piston area of ​​the hydraulic cylinder, the flow-opening characteristics of the valve orifice, and the transmission ratio of the counter-attack plate, ensuring a linear and decoupled relationship between the control quantity and the actual speed. Simultaneously, the directional information is determined by the sign of the adjustment error data: a positive sign indicates an increase in clearance, and a negative sign indicates a decrease in clearance. To balance actuator safety and overall machine stability, the controller superimposes soft limiting and acceleration shaping on the target hydraulic speed. First, the target speed is proportionally converted to form the desired value, then double-trimmed using speed and acceleration thresholds: when the target hydraulic speed exceeds the preset speed threshold, the hydraulic cylinder's action speed is limited to the preset speed threshold; in cases where the speed is within the limit but the transition is too rapid, the slope of the speed change is constrained by the maximum increment per cycle. The speed protection parameter is a maximum adjustment speed of 2 mm / s, and the speed threshold is set to 2 mm / s. After speed limiting, the limited hydraulic speed is obtained. The controller interpolates the limited hydraulic speed with the actual speed of the previous cycle to form a speed trajectory with S-shaped transition characteristics. Simultaneously, a small dead zone and low-gain softening are introduced in the small deviation range to allow the speed to naturally decay as it approaches the target gap, avoiding high-frequency back-and-forth fine-tuning within 1mm. The controller generates hydraulic adjustment commands based on the limited hydraulic speed. These commands include speed setting, direction position, maximum allowable valve opening, and the expected displacement increment for the current cycle, along with a timestamp for execution side tracking.

[0035] In one specific embodiment, the correction subsystem further includes: The acquisition module is used to measure the difference between the current temperature of the motor and the initial temperature at startup to obtain the motor temperature rise data, and at the same time, to determine the current wear status of the impact plate based on the motor temperature rise data; The calculation module is used to calculate the wear gap deviation of the impact plate based on the current wear condition; The prediction module is used to analyze the real-time change rate of the target load current, multiply the real-time change rate by a preset prediction coefficient and a delay time constant to predict the time difference between the hydraulic cylinder receiving the first gap adjustment command and completing the action, and obtain the hydraulic response delay. The compensation module is used to add the wear clearance deviation to the first clearance adjustment command to compensate for the wear effect, and then adjust the command timing in advance according to the hydraulic response delay to obtain the second clearance adjustment command.

[0036] Specifically, the acquisition module collects real-time temperature data of the main drive motor during the operation of the forming machine. It uses a temperature sensor to obtain the current operating temperature and records the initial temperature as a baseline value upon system startup. The motor's temperature rise data is obtained by subtracting the initial startup temperature from the current temperature. This temperature rise data reflects the motor's heat generation level under current load conditions and indirectly characterizes the wear state of the impact plate. Because the impact plate's working resistance increases when processing high-hardness materials or encountering high impact frequencies over long periods, the motor operates under high load for extended periods, causing additional heat generation. The acquisition module compares the temperature rise data with historical operating data and calibration curves. Using an empirical model between temperature rise and wear degree, it determines the current wear level of the impact plate. For example, if the motor temperature rise exceeds 80% of the rated temperature rise, the wear state is considered significant, triggering a corresponding compensation strategy. The calculation module calculates the wear gap deviation of the impact plate based on the wear state. The calculation method combines parameters such as historical wear coefficient, operating time, and load intensity. For example, the initial value of the wear compensation coefficient is set to 0.001 mm / (A·h), and it is increased by 20% when the temperature rise exceeds the limit, reflecting the deviation of the actual gap of the impact plate relative to the set value under different wear levels. The calculation module uses the current cumulative operating current integral and the temperature rise correction coefficient to obtain the additional gap change caused by wear, i.e., the wear gap deviation. A positive value indicates an increase in gap, and a negative value indicates a decrease in gap. The prediction module predicts the execution delay based on the dynamic response characteristics of the hydraulic actuator. It reads the current target load current and calculates its rate of change in real time, reflecting the speed of change in the load state of the impact plate. The prediction module multiplies the real-time rate of change with the preset prediction coefficient and combines it with the delay time constant of the hydraulic system to estimate the time difference between the hydraulic cylinder receiving the first gap adjustment command and completing the physical action, i.e., the hydraulic response delay. The prediction coefficient is set to 0.8, and the delay time constant is 0.2 seconds. The compensation module directly adds the wear clearance deviation to the target clearance corresponding to the first clearance adjustment command, thereby compensating for the clearance error caused by the wear of the impact plate and ensuring that the impact plate can maintain a working distance close to the design value after long-term operation. Based on the hydraulic response delay given by the estimation module, the compensation module pre-adjusts the execution timing of the first clearance adjustment command, allowing the hydraulic system to start operation in advance. After delay compensation, the impact plate can accurately reach the required clearance position at the target time. After this dual compensation process, the compensation module generates the second clearance adjustment command.

[0037] In one specific embodiment, the calculation module is specifically used for: Calculate the percentage of wear exceeding the standard wear threshold from the current wear condition; The first wear compensation coefficient is adjusted based on the proportion of wear exceeding the standard to obtain the second wear compensation coefficient; Accumulate the load current values ​​of the shaping machine from its first start-up to the current moment for all operating periods, and continuously integrate the load current values ​​to record the total impact load borne by the shaping machine, thereby obtaining the historical integrated value of the load current. The wear gap deviation caused by cumulative impact wear of the impact plate is quantified by multiplying the second wear compensation coefficient by the historical integral value of the load current.

[0038] Specifically, a comparison mechanism is established between real-time wear status and standard wear threshold. The current wear status is quantified into a dimensionless wear degree index, for example, using a ratio of 0 to 1 to represent the degree of wear progression relative to the design life. The standard wear threshold is obtained through equipment design parameters and experimental calibration, characterizing the maximum wear level that the impact plate can withstand before reaching the specified maintenance cycle under normal operating conditions. The difference between the current wear status and the standard wear threshold is calculated, and the difference is divided by the standard wear threshold to obtain the wear excess ratio, which represents the percentage of the current wear level exceeding the normal safe range. A positive value represents exceeding the standard, while a zero or negative value represents within the safe range. The first wear compensation coefficient is dynamically adjusted according to the wear excess ratio to obtain the second wear compensation coefficient. The first wear compensation coefficient is a benchmark compensation value set at the factory or after major overhaul, set to 0.001 mm / (A·h), used to convert the cumulative current impact amount into the gap change. If the wear excess ratio is high, it indicates that the impact plate is in an accelerated deterioration stage, so the compensation coefficient needs to be increased to increase the correction force; if the wear level is low, the compensation coefficient is maintained or appropriately reduced to avoid over-compensation causing control errors. In actual calculations, the adjustment process is achieved through linear proportional correction or piecewise functions. For example, when the wear exceedance ratio is 0.2, the first wear compensation coefficient is multiplied by (1+0.2) to obtain a new second wear compensation coefficient. The load current values ​​of the shaping machine during all operating periods from its initial startup to the current moment are accumulated, and a quantitative index reflecting the total stress history of the equipment is obtained through continuous integration. Instantaneous load current values ​​are collected within the control cycle, and these instantaneous values ​​are integrated over time. The integration process needs to ensure data continuity; therefore, the interval between equipment stops and starts is marked to ensure that the integration calculation covers all effective operating periods without repeatedly counting non-operating states, thus obtaining the total impact load, reflecting the total energy and stress impact borne by the impact plate throughout its entire lifespan. The accumulated impact energy is converted into the actual wear gap deviation of the impact plate by multiplying the second wear compensation coefficient by the historical load current integration value, representing the gap change caused by long-term impact processing.

[0039] In one specific embodiment, the control system of the forming machine for recycled aggregate from construction waste further includes a blockage risk analysis subsystem, which includes: The differential operation module is used to perform two consecutive differential operations on the target load current to obtain the second derivative. The processing module is used to collect load current fluctuation data, calculate the standard deviation and root mean square deviation of the load current fluctuation data, and calculate the load current variance value based on the standard deviation and root mean square deviation. The synthesis module is used to synthesize the second derivative and the load current variance value into blockage precursor identification parameters, and generate a blockage risk value based on the blockage precursor identification parameters; The risk control module is used to increase the gap value in the second gap adjustment command to a safe gap value when the material blockage risk value exceeds the risk threshold, and to keep the second gap adjustment command unchanged when the material blockage risk value does not exceed the risk threshold, and finally obtain the third gap adjustment command.

[0040] Specifically, the differential operation module performs two consecutive differential operations on the target load current according to the time series. The first differential yields the rate of change of current, used to characterize the instantaneous speed of load change over time. The second differential yields the second derivative, characterizing the rate of change of current itself, i.e., the acceleration characteristic. The processing module collects load current fluctuation data in parallel, extracting fluctuation data from the target load current within a fixed-length time window (e.g., 5 seconds). The processing module calculates the standard deviation of the load current samples within the time window to measure the amplitude of current fluctuations and calculates the root mean square deviation to reflect the energy level of current fluctuations. The standard deviation and root mean square deviation reveal the stability of the current load signal in different dimensions. When both increase simultaneously, it indicates that the system is in an unstable state affected by external disturbances or material accumulation. The processing module combines the standard deviation and root mean square deviation to calculate the variance value of the load current. The synthesis module merges the second derivative output by the differential operation module with the load current variance value output by the processing module to form the pre-blockage identification parameters. The synthesis process employs weighted superposition or normalized multiplication to ensure the comparability of the second derivative and variance values ​​on a numerical scale. The second derivative is used to capture the acceleration characteristics of the blockage trend, while the variance value is used to verify the persistence and global stability of the blockage trend. The synthesized blockage precursor identification parameters exhibit high sensitivity and robustness to different blockage formation processes (e.g., sudden foreign object blockage and gradual accumulation blockage). Based on this, the synthesis module performs threshold mapping or nonlinear function transformation on the blockage precursor identification parameters, converting them into blockage risk values ​​ranging from 0 to 1 or expressed as a percentage. The risk control module dynamically adjusts the gap setting of the second gap adjustment command based on the comparison between the blockage risk value and the preset risk threshold. When the blockage risk value exceeds the risk threshold, it indicates that the current material flow state has approached or entered a high-risk blockage stage. At this point, the risk control module immediately intervenes, increasing the gap value in the second gap adjustment command to a safe gap value, thereby reducing resistance and material compaction density within the crushing chamber, allowing material to pass through more smoothly and reducing further blockage development. The safety clearance value is calibrated experimentally during the equipment commissioning phase. If the risk value of material blockage does not exceed the risk threshold, it indicates that the current working state is still within an acceptable range. The risk control module maintains the second clearance adjustment command unchanged to preserve the original production efficiency and energy consumption optimization strategy. After this decision, the module outputs an updated third clearance adjustment command.

[0041] In one specific embodiment, the synthesis module is specifically used for: The sensitivity weighting coefficient of the second derivative and the fluctuation weighting coefficient of the load current variance are determined based on historical material blockage events. Multiplying the second derivative by the sensitivity weighting coefficient yields the first blockage parameter, and multiplying the load current variance by the fluctuation weighting coefficient yields the second blockage parameter. Construct positive and negative sample pairs for the twin support vector machine, use the identification parameters during historical normal operation as negative samples, and use the identification parameters during historical material blockage precursors as positive samples, train the twin support vector machine to learn the feature boundaries between the material blockage state and the non-material blockage state, and obtain the material blockage identification classifier; The first and second blockage parameters are used as blockage precursor identification parameters and input into the blockage identification classifier. The similarity distance between the blockage precursor identification parameters and positive samples and the similarity distance with negative samples are calculated, and the blockage risk value is output.

[0042] Specifically, a weighted adaptive calibration process is established based on historical material blockage events. The period of 5–30 seconds preceding a blockage is extracted from archived operation logs as a "precursor segment," and periods matching the power level, feed rate, and material type without anomalies are extracted as "normal segments." Both are sampled at 1000Hz and a 5-second sliding window is used to generate feature sequences. Within each window, the second derivative of the current output by the differential calculation module is used to characterize accelerated abrupt changes, and the variance of the load current output by the processing module is used to measure the overall fluctuation intensity. A weight search aimed at maximizing AUC or achieving F1 optimization is used to jointly optimize the sensitive weight coefficients of the second derivative and the fluctuation weight coefficients of the variance. The optimal strategy employs grid search or Bayesian optimization, and cross-domain time validation is used to avoid overfitting to a single operating condition. During initialization, the weights of the variance channel are aligned with the historical variance weight coefficient of 0.3 as the starting point, and the optimizer is then automatically fine-tuned within the neighborhood. When equipment models or material composition change significantly, the system triggers recalibration in batches, allowing weights to adapt to changes in operating conditions. After determining the weights, the second derivative is multiplied by the sensitive weight to obtain the first blockage parameter, and the variance is multiplied by the fluctuation weight to obtain the second blockage parameter. Both are then standardized to zero mean and scaled to avoid skewness in the classifier's decision surface caused by differences in dimensions. During the model training phase, positive and negative sample pairs are constructed for the twin support vector machine. Negative samples are derived from the blockage precursor identification parameter sequence extracted from "normal segments," and positive samples are derived from the identification parameter sequence extracted from "precursor segments." To avoid class imbalance, resampling is performed before training based on the number of windows or the hard case mining ratio, and clustering downsampling is performed on highly similar redundant windows to improve the sparsity of the discrimination boundary. The twin support vector machine simultaneously learns two parallel or nearly parallel hyperplanes, minimizing the distance of each class's sample point to its own hyperplane and maximizing its distance to the other class's hyperplane, resulting in two mutually constrained classification boundaries. In engineering practice, linear kernels are preferred for high interpretability and real-time performance. When the operating data exhibits a significant nonlinear distribution, a radial basis kernel is used with a limited kernel width to prevent overfitting in volatile but non-blocking conditions. The training process employs hierarchical time-segmented cross-validation, dividing the entire historical timeline into blocks by date or batch to ensure the validation set covers different materials and load ranges. In addition to AUC and F1, the evaluation metrics also track the "lead time - accuracy" trade-off curve to ensure the classifier not only makes correct judgments but also identifies problems earlier. If false alarms are found to be concentrated in the recovery phase after blockage removal on the validation set, short-term energy fallback indicators (e.g., the negative pulse ratio of the second derivative) are added as an auxiliary dimension to the training features, while the first and second blockage parameters remain the primary features to maintain simplicity and efficient inference. After training, a blockage identification classifier is obtained. During online inference, the synthesis module generates the first and second blockage parameters in each control cycle and inputs them into the classifier.The twin support vector machine calculates the signed distances from two-dimensional feature points to the positive and negative hyperplanes. These distances are then transformed into similarity distances with positive and negative samples through a gentle monotonic mapping (such as Sigmoid or Logit calibration). The blockage risk value is output using normalized difference or soft maximization, stabilizing it between 0 and 1. A higher value indicates a closer approximation to historical precursor patterns. A risk value time stabilizer is added: when the risk suddenly increases within a single cycle but does not continue to rise in adjacent cycles, the risk is slightly pulled back by the suppression coefficient to avoid false alarms triggered by isolated noise; when the risk rises continuously and exceeds the empirical trigger zone, the risk is appropriately amplified for earlier intervention.

[0043] The process involves synthesizing the second derivative of the load current and the variance of the load current into a parameter for identifying the precursors of material blockage, weighted according to a preset ratio. This includes: establishing a precursor identification operator; constructing a composite identification parameter through the additive fusion of the second derivative and variance terms of the load current, where the second derivative term captures the acceleration mutation characteristics of the load current, and the variance term quantifies the random fluctuation characteristics of the load current, resulting in a second derivative-variance fusion operator; designing a numerical differentiation algorithm for the second derivative term, employing a five-point central difference scheme to calculate the second derivative of the load current to avoid noise amplification effects in numerical differentiation; smoothing the second derivative sequence using a Savitzky-Golay filter; setting the second derivative anomaly threshold to 50 amperes per square second; and constructing a sliding window statistical algorithm for the variance term. The load current variance is calculated using a 5-second sliding window. Bessel correction is used to improve the unbiasedness of small-sample variance estimation. A variance weighting coefficient of 0.3 is set to balance the contribution of the second derivative and variance, resulting in a variance feature extractor. An adaptive threshold determination mechanism for the identification operator is established. The optimal identification threshold is determined through ROC curve analysis. Historical blockage event data is used to train the threshold parameters, and a balanced optimization model for false positives and false negatives is established. The discrimination threshold of the identification operator is dynamically adjusted, resulting in an adaptive threshold setter. A joint time-domain and frequency-domain analysis of the identification operator is designed. Wavelet transform is performed on the identification operator sequence to analyze its frequency domain characteristics, identifying the characteristic frequency bands and duration patterns of blockage precursors. A joint discrimination criterion of time-domain statistical characteristics and frequency-domain energy distribution is established, resulting in a multi-domain fusion blockage identification system.

[0044] In this invention, real-time automatic identification of construction waste composition is achieved through motor load current characteristic analysis. This accurately distinguishes materials of different hardness, such as concrete blocks, waste bricks, and stone, and effectively eliminates interference from lightweight impurities such as insulation materials and plastics, solving the technical problem of traditional technologies being unable to identify material differences. Based on the dynamic selection of dedicated optimization coefficients according to material composition type, the optimal balance between shaping efficiency and energy consumption is calculated through a three-dimensional relationship function. The gap of the impact plate is automatically adjusted according to the characteristics of different construction wastes, significantly improving the stability and consistency of the shaping effect compared to traditional fixed-gap control methods. A segmented PID control strategy is adopted, automatically switching control parameters according to the gap error magnitude. Combined with a dedicated stability criterion to prevent positive feedback oscillation between the gap and load, a dynamic control mode with fast response, balanced adjustment, and fine control is achieved, effectively solving the overshoot and oscillation problems in traditional control. The wear state of the impact plate is evaluated in real time using motor temperature rise data, and the wear compensation coefficient is dynamically adjusted. Combined with historical load data throughout the equipment's life cycle, the gap deviation caused by wear is accurately calculated, ensuring the stability of control accuracy during long-term operation and overcoming the technical deficiency of traditional technologies that ignore the impact of wear. Based on the prediction of hydraulic system response delay using the load current change rate, the command timing is adjusted in advance for compensation and correction, eliminating the impact of system lag on control accuracy and achieving more precise and real-time gap adjustment response. A blockage precursor identification system is constructed using the second derivative of the load current and variance analysis, enabling early warning and proactive prevention of blockage risks, thus improving the continuous operation capability and reliability of the forming machine.

[0045] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and units described above can be referred to the corresponding processes in the foregoing system embodiments, and will not be repeated here.

[0046] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the system described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0047] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A control system for a shaping machine for recycled aggregate from construction waste, characterized in that, include: The acquisition subsystem is used to acquire the target load current of the main drive motor in the shaping machine and identify the material composition type based on the target load current; A calculation subsystem is used to select the optimization coefficient corresponding to the material composition type and calculate the target gap based on the optimization coefficient; The control subsystem is used to generate adjustment error data based on the target gap and the current gap, and to control the hydraulic cylinder to adjust the position of the counterattack plate according to the adjustment error data, thereby generating a first gap adjustment command. The correction subsystem is used to detect motor temperature rise data and predict hydraulic response delay, and correct the first gap adjustment command based on the hydraulic response delay to obtain the second gap adjustment command.

2. The control system of the shaping machine for recycled aggregate from construction waste according to claim 1, characterized in that, The acquisition subsystem also includes: The acquisition module is used to acquire the instantaneous load current of the main drive motor in the shaping machine; The identification module is used to measure the duration of the peak signal and the current drop slope in the instantaneous load current, identify and eliminate interference signals from lightweight materials, and obtain the target load current. The analysis module is used to divide the target load current by the motor rated current to obtain the load current ratio. The identification module is used to identify the material composition type as concrete block when the load current ratio is greater than a first target value; to identify the material composition type as waste brick when the load current ratio is between a second target value and a first target value; and to identify the material composition type as stone when the load current ratio is less than a second target value, wherein the first target value is greater than the second target value.

3. The control system of the shaping machine for recycled aggregate from construction waste according to claim 2, characterized in that, The identification module is specifically used for: Identify the peak segments in the instantaneous load current that exceed the average load current, and measure the time span from the start of the rise to the end of the fall for each peak segment to obtain the duration; At the end of each peak period, calculate the rate of change of the load current from the peak to the average value to obtain the current drop slope. The duration is compared with the time discrimination threshold, and the current drop slope is compared with the slope discrimination threshold. When the duration is shorter than the time discrimination threshold and the current drop slope exceeds the slope discrimination threshold, it is marked as a lightweight material interference signal. The current data marked as interference signals of lightweight materials are removed from the instantaneous load current, and the effective load characteristics corresponding to heavy construction waste are retained to obtain the target load current.

4. The control system of the shaping machine for recycled aggregate from construction waste according to claim 1, characterized in that, The computing subsystem is specifically used for: According to the material composition type, the corresponding optimization coefficient is retrieved from the preset coefficient library, where concrete blocks correspond to the first set of coefficients, waste bricks correspond to the second set of coefficients, and stone corresponds to the third set of coefficients. The current load current of the shaping machine and the residence time of the material in the cavity are collected synchronously to obtain the current operating status parameters; The optimization coefficients are used to construct a relationship function between the counter-attack plate gap and the shaping efficiency and energy consumption. The current operating state parameters are substituted into the relationship function to calculate the response value. The extreme points are determined based on the response values, and the target gap is calculated based on the extreme points to maximize the shaping efficiency while minimizing energy consumption.

5. The control system of the shaping machine for recycled aggregate from construction waste according to claim 1, characterized in that, The control subsystem also includes: The judgment module is used to subtract the current gap of the counterattack plate from the target gap to obtain the gap deviation, and at the same time determine the absolute value range of the gap deviation to obtain adjustment error data; The selection module is used to select the corresponding current PID parameters based on the adjustment error data. The generation module is used to generate hydraulic adjustment commands based on the adjustment error data and the current PID parameters; The verification module is used to verify whether the hydraulic adjustment command satisfies the stability condition that the product of the gap change gradient and the load change gradient is less than zero. When the stability condition is met, the hydraulic adjustment command is used as the first gap adjustment command and output.

6. The control system of the shaping machine for recycled aggregate from construction waste according to claim 5, characterized in that, The generation module is specifically used for: The adjustment error data is multiplied by the proportional coefficient in the current PID parameters to obtain the proportional control component. The adjustment error data is integrated over time and then multiplied by the integral coefficient to obtain the integral control component. The adjustment error data is differentiated over time and then multiplied by the derivative coefficient to obtain the derivative control component. The proportional control component, the integral control component, and the derivative control component are proportionally converted to obtain the target hydraulic speed. When the target hydraulic speed exceeds a preset speed threshold, the hydraulic cylinder's operating speed is limited to the preset speed threshold; when the target hydraulic speed does not exceed the preset speed threshold, the target hydraulic speed remains unchanged, thus obtaining the limited hydraulic speed. Hydraulic adjustment commands are generated based on the restricted hydraulic speed.

7. The control system of the shaping machine for recycled aggregate from construction waste according to claim 1, characterized in that, The correction subsystem also includes: The acquisition module is used to measure the difference between the current temperature of the motor and the initial temperature at startup to obtain the motor temperature rise data, and at the same time, to determine the current wear status of the impact plate based on the motor temperature rise data; The calculation module is used to calculate the wear gap deviation of the impact plate based on the current wear state; The prediction module is used to analyze the real-time change rate of the target load current, multiply the real-time change rate by a preset prediction coefficient and a delay time constant to predict the time difference between the hydraulic cylinder receiving the first gap adjustment command and completing the action, and obtain the hydraulic response delay. The compensation module is used to add the wear gap deviation to the first gap adjustment command to compensate for the wear effect, and then adjust the command timing in advance according to the hydraulic response delay to obtain the second gap adjustment command.

8. The control system of the shaping machine for recycled aggregate from construction waste according to claim 7, characterized in that, The calculation module is specifically used for: Calculate the percentage of wear exceeding the standard wear threshold compared to the current wear state; The first wear compensation coefficient is adjusted based on the wear exceeding the standard ratio to obtain the second wear compensation coefficient; Accumulate the load current values ​​of the shaping machine from its first start to the current moment for all operating periods, and continuously integrate the load current values ​​to record the total impact load borne by the shaping machine, thereby obtaining the historical integrated value of the load current. The wear gap deviation caused by cumulative impact wear of the impact plate is quantified by multiplying the second wear compensation coefficient by the historical integral value of the load current.

9. The control system of the shaping machine for recycled aggregate from construction waste according to claim 1, characterized in that, The control system of the forming machine for recycled aggregate from construction waste also includes a blockage risk analysis subsystem, which includes: The differential operation module is used to perform two consecutive differential operations on the target load current to obtain the second derivative. The processing module is used to collect load current fluctuation data, calculate the standard deviation and root mean square deviation of the load current fluctuation data, and calculate the load current variance value based on the standard deviation and root mean square deviation. The synthesis module is used to synthesize the second derivative and the load current variance value into a blockage precursor identification parameter, and generate a blockage risk value based on the blockage precursor identification parameter; The risk control module is used to increase the gap value in the second gap adjustment command to a safe gap value when the material blockage risk value exceeds the risk threshold, and to keep the second gap adjustment command unchanged when the material blockage risk value does not exceed the risk threshold, and finally obtain the third gap adjustment command.

10. The control system of the shaping machine for recycled aggregate from construction waste according to claim 9, characterized in that, The synthesis module is specifically used for: The sensitivity weighting coefficient of the second derivative and the fluctuation weighting coefficient of the load current variance are determined based on historical material blockage events. Multiplying the second derivative by the sensitivity weighting coefficient yields the first blockage parameter, and multiplying the load current variance by the fluctuation weighting coefficient yields the second blockage parameter. Construct positive and negative sample pairs for the twin support vector machine, use the identification parameters during historical normal operation as negative samples, and use the identification parameters during historical material blockage precursors as positive samples, train the twin support vector machine to learn the feature boundaries between the material blockage state and the non-material blockage state, and obtain the material blockage identification classifier; The first and second blockage parameters are used as blockage precursor identification parameters and input into the blockage identification classifier. The similarity distance between the blockage precursor identification parameters and positive samples and the similarity distance with negative samples are calculated, and the blockage risk value is output.

Images